CS emission near MIR-bubbles C. Watson Manchester University, Dept. of Physics, 604 E. College Ave., North Manchester, IN 46962 6 1 0 [email protected] 2 n a Kathryn Devine J 5 College of Idaho, Dept. of Physics, 2112 Cleveland Blvd, Caldwell, ID 83605 ] A G [email protected] . h p N. Quintanar - o r Texas A&M University, Dept. of Nuclear Engineering, 401 Joe Routt Blvd, College t s a Station, TX 77843 [ 1 v [email protected] 6 0 9 0 T. Candelaria 0 . New Mexico Institute of Mining and Technology, Dept. of Physics, 801 Leroy Place, 1 0 6 Socorro, NM 87801 1 : v i [email protected] X r a Received ; accepted – 2 – ABSTRACT We survey 44 young stellar objects located near the edges of mid-IR-identified bubbles in CS (1-0) using the Green Bank Telescope. We detect emission in 18 sources, indicating young protostars that are good candidates for being triggered by the expansion of the bubble. We calculate CS column densities and abun- dances. Three sources show evidence of infall through non-Gaussian line-shapes. Two of these sources are associated with dark clouds and are promising candi- dates for further exploration of potential triggered star formation. We obtained on-the-fly maps in CS (1-0) of three sources, showing evidence of significant in- teractions between the sources and the surrounding environment. Subject headings: stars: formation, ISM: HII regions, ISM: molecules, radio lines: ISM – 3 – 1. Introduction Prior to post-main-sequence evolution, ionizing radiation is one of the most important mechanisms by which massive stars influence their surrounding environments. This ionizing radiation may potentially trigger subsequent star-formation. The influence of ionizing radiation is observed in the form of bubble-shaped emission in the 8 µm band of the Spitzer-GLIMPSE survey of the Galactic Plane (Benjamin et al. 2003). Churchwell et al. (2006, 2007) observed bubble-shaped 8 µm emission to be common throughout the Galactic plane. Watson et al. (2008, 2009) found 24 µm and 20 cm emission centered within the 8 µm emission and interpreted the bubbles seen in the GLIMPSE data as caused by hot stars ionizing their surroundings, creating 20 cm free-free emission, and at larger distances exciting PAHs, creating 8 µm emission. Deharveng et al. (2010) also interpreted the bubbles as classical HII regions. Watson et al. (2010) used 2MASS and GLIMPSE photometry and Spectral Energy Distribution (SED)-fitting to analyze the YSO population around 46 bubbles and found about a quarter showed an overabundance of YSOs near the boundary between the ionized interior and molecular exterior. These YSOs are candidates for being triggered by the expanding ionization and shock fronts created by the hot star. Star formation triggered by previous generations of stars is known to occur but the specific physical mechanism is still undetermined. The collect-and-collapse model (Elmegreen & Lada 1977) describes ambient material swept up by the shock fronts which eventually becomes gravitationally unstable, resulting in collapse. Other mechanisms, however, have been proposed. Radiatively-driven implosion (Lefloch & Lazareff 1994), for example, describes clumps already present in the ambient material whose contraction is aided by the external radiation of the hot star. Bubbles with an overabundance of YSOs along the bubble-interstellar medium (ISM) boundary are a potentially excellent set of sources to study the mechanisms of triggered – 4 – star-formation. The method of identifying YSOs through photometry, however, is limited. Robitaille et al. (2006) showed that YSO age is degenerate with the observer’s inclination angle. An early-stage YSO and a late-stage YSO seen edge on, so the accretion or debris disk is observed as thick and blocking the inner regions, can appear similar, even in the IR. Thus, we require other diagnostics of the YSOs along the bubble edge to determine the youngest, and most likely to have been triggered, YSOs. Additionally, a line-diagnostic allows us to rule out any line-of-sight coincidence associations. For the current project we selected a subset of the bubbles identified above to identify those YSOs associated with infall, outflows or hot cores by observing the CS (1-0) transition near 49 GHz with the Green Bank Telescope (GBT1). CS is a probe of young star-formation. It has been detected in outflows from protostars, infall, disks and in hot cores (Dutrey et al. 1997; Bronfman et al. 1996; Morata et al. 2012). The chemistry is, naturally, complex, and it appears that CS can play several roles (Beuther et al. 2002), such as tracing outflows (Wolf-Chase et al. 1998) or hot cores (Chandler & Wood 1997). Our aim here is to use CS as a broad identifier of young star-formation and use any non-Gaussian line-shapes to infer molecular gas behavior. After describing the CS survey and CS mapping observations (§ 2) and numerical results (§ 3), we analyze the Herschel-HiGAL emission toward all our sources to determine, along with our CS detections, the CS abundances (§ 4.1). We also analyze three sources with evidence of rapid infall (§ 4.2). We end with a summary of the conclusions. 1The National Radio Astronomy Observatory is a facility of the National Science Foun- dation operated under cooperative agreement by Associated Universities, Inc. – 5 – 2. Observations Candidate YSO locations were identified using the SED fitter tool developed by Robitaille et al. (2006, 2007). Briefly, this method uses the 2MASS (Kleinmann et al. 1994) and GLIMPSE point source catalogues to identify sources that are not well-fit by main-sequence SEDs and are well-fit by YSO SEDs. Watson et al. (2010) fit all point sources within 1′of the bubble edges using this method. From this set of point sources, four sources were selected near each bubble based on association with either diffuse, bright 8 µm emission or IR dark clouds. Forty point sources in total were selected. The names, Galactic longitude and Galactic latitude are reported in Table 2. Each point source was observed for CS using the Green Bank Telescope (GBT) for two 5 minute integrations. The spectrometer was set-up in frequency switching mode to maximize on-source observing time. The setup parameters and calibration sources are listed in Table 1. Data were calibrated and analyzed using GBTIDL. Typical system temperatures were between 105 K and 120 K. Typical rms noise in the resulting calibrated spectra was 0.20 K. Non-detections and detections are listed in Tables 2 and 3, respectively. We estimate uncertainty due to flux calibration of 20%. In addition to single pointings, we mapped three regions (N56, N65 2 and N77 1) that displayed strong CS emission. The map sizes were 1′x1 ′(N56 and N77-1) and 2′x2′(N65-2), both using a Nyquist-sampling step-size of 6.12′′. Observations were used from the Hi-Gal (Molinari et al. 2010) project, a Herschel Space Telescope imaging survey of the Galactic plane. This survey observed all the sources in this study at wavelengths between 60 µm and 600 µm. Data were downloaded from the Spitzer Science Center website. Level 2 data products were used, which have been fully calibrated. – 6 – Table 1. Observing Parameters Bandwidth 50 MHz Channel width 1.5 kHz Rest frequency 48.99095 GHz Frequency switching shift 8 MHz Pointing calibration 1751+0930 1850-0001 2025+3343 Flux calibration NGC7027 – 7 – Table 2. CS Non-Detections Name l(◦) b(◦) N62-2 34.329 0.195 N62-3 34.317 0.197 N65-3 34.963 0.310 N65-4 35.049 0.330 N77-3 40.407 -0.037 N77-4 40.409 -0.033 N82-1 42.122 -0.635 N82-2 42.128 -0.636 N82-3 42.114 -0.616 N82-4 42.112 -0.658 N90-3 43.748 0.0754 N90-4 43.735 0.0629 N92-1 44.359 -0.825 N92-4 44.335 -0.824 N117-1 54.102 -0.094 N117-2 54.076 -0.085 N123-1 57.562 -0.297 N123-3 57.567 -0.285 N123-4 57.564 -0.280 N128-1 61.688 0.990 N128-2 61.703 0.988 – 8 – 3. Results 3.1. CS Point Sources Eighteen sources displayed emission greater than 3σ. A typical spectrum is shown in Figure 1. Emission lines were fit using fitgauss, the standard Gaussian fitting routine in GBTIDL. Fitting parameters (amplitude in T units, central velocity and FWHM) are mb listed in Table 3.1. For sources that displayed a double peak, two simultaneous Gaussian functions were fit to the emission and are listed in consecutive rows. CS column densities, N , were calculated assuming LTE, optically thin emission and an excitation temperature CS T =15 K, a typical ISM value (see review in Zinnecker & Yorke (2007)). Increasing or ex decreasing the assumed excitation temperature by 5 K changes the column density by about 30%. If CS(1-0) is optically thick, as we assume for three sources in section 4.2 below, then our calculation would be a lower limit. Given these assumptions we used the following relation (see Miettinen (2012) for a detailed discussion of the relations below): 3k ǫ 1 Z (T ) eEu/kBTex B 0 rot ex N = T dv CS 2π2 νµ2elS gKgI 1− F(Tbg) Z MB F(Tex) where Table 2—Continued Name l(◦) b(◦) N128-3 61.625 0.953 N128-4 61.704 0.921 – 9 – g = g = 1 K I µ2S = 3.8 Debye2 el Z = 0.8556 T −0.10 rot ex F(T) = 1 . ehν/kBT−1 Here ǫ is the vacuum permittivity, µ is the permanent electric dipole moment, S is 0 el the line strength, Z is the rotational partition function, ν is the frequency, g is the rot K K-level degeneracy, g is the reduced nuclear spin degeneracy, E is the energy of the I u upper-transition state, T is the excitation temperature and T is the background ex bg temperature. The dipole moment line strength (µ2 S) is taken from the JPL spectral el line catalog (Pickett et al. 1998). The partition function (Z ) is a linear fit to JPL data rot between T=37 to 75 K. T was taken to be the cosmic microwave background temperature, bg 2.725 K. The uncertainty in the fit amplitudes and derived column densities is dominated by our flux-calibration uncertainty. Since the relationships are linear, we estimate the uncertainty in both as 20%. – 10 – Table 3. Gaussian fitting parameters for CS detections. Name l (◦) b (◦) T (K) Vel (km/s) FWHM (km/s) N (cm−2) mb LSR CS N62-1 34.352 0.192 1.3 57.6 0.9 3.4×1014 3.5 56.4 1.2 1.3×1015 N62-2 34.329 0.195 1.1 57.2 1.9 6.1×1014 N65-1 35.044 0.327 1.6 51.3 1.7 8.1×1014 N65-2 35.025 0.350 2.4 50.4 2.2 1.6×1015 10.9 53.3 4.1 1.3×1016 N65-4 35.049 0.330 2.2 51.3 1.9 1.1×1015 N77-1 40.437 -0.044 3.1 68.0 1.6 1.4×1015 N77-2 40.422 -0.024 4.1 69.5 2.6 2.9×1015 N82-5 42.125 -0.623 5.7 66.4 1.8 2.9×1015 N90-1 43.788 0.083 1.2 35.2 0.7 2.1×1014 N90-2 43.792 0.089 1.0 36.0 0.6 1.4×1014 2.6 35.4 0.6 3.9×1014 N92-2 44.349 -0.803 1.3 61.6 1.4 5.8×1014 N92-3 44.334 -0.818 1.9 61.3 2.6 1.5×1015 N117-3 54.107 -0.044 2.3 40.9 1.7 1.2×1015 3.8 38.4 2.3 2.7×1015 N123-2 57.578 -0.284 2.3 -9.3 1.9 1.4×1015 N133-1 63.125 0.442 1.0 20.7 2.2 7.0×1014 N133-2 63.132 0.415 1.4 19.3 2.2 7.3×1014 N133-3 63.179 0.440 1.4 23.3 2.2 7.4×1014